His team is using artificial intelligence to help kill cancer

A robotic arm developed by researchers at the University of Waterloo aims to help doctors more efficiently kill cancerous tumours using artificial intelligence.

It works by helping guide doctors to more accurately focus high-intensity focused ultrasound treatment or HIFU on cancerous tumours, limiting the amount of damage done to healthy, non-cancerous cells.

Moslem Sadeghi Goughari, the research lead behind the project, said the robotic arm is the first to use artificial intelligence in this way.

"Artificial intelligence right now is an accepted technology for healthcare. It's currently used, for example, for cancer diagnosis. But in terms of the HIFU monitoring, we are the first group of people that's using A.I. for this procedure," he said.

Here's how this robotic arm uses AI to kill cancer

1 year ago

Duration 3:00

Researchers at the University of Waterloo have created a robotic arm with artificial intelligence that is trained to kill cancer. In this video, the project's lead researcher, Moslem Sadeghi Goughari, demonstrates how the robotic arm works.

He said his team has developed the first AI-powered ultrasound imaging system for non-invasive cancer treatment.

HIFU is a minimally invasive medical procedure that is used to treat tumours using ultrasound waves that heat and kill cancer cells.

The robotic arm that Sadeghi Goughari and his team created will allow doctors to more accurately monitor the progress of HIFU treatment as it is happening. This is made possible by equipping the robotic arm with artificial intelligence.

"Ultrasound imaging [in HIFU] is not easy for doctors. They need higher skills, they need training and it's not really accurate because it depends on the skill of the doctor if they can detect the updated area," he said.

"Technically, it's considered the biggest barrier for the HIFU treatment."

Sadeghi Goughari said because the robotic arm reduced damage to non-cancerous, healthy cells, patients will have a shorter recovery time of two days, compared to the average of 10 days needed to recover after each HIFU treatment session.

Potential downsides of using AI

Sadeghi Goughari said there are some downsides to using AI because the accuracy of the robot depends on the data used to train it.

"That's why at this stage, in our lab, we're using this technology on the in-vitro medium. It's not on live tissue," he said.

To fix that, the team is working on getting more accurate data from a doctor and researcher they have partnered with in South Korea.

The next step after that is to do more testing and receive FDA approval before the robot is available for cancer treatment at local hospitals.

"The problem we are solving is the problem that everyone in ... the field of non-invasive treatment is talking about," Sadeghi Goughari said.